Skip to main content

    Listening to machines to understand why they break

    Listening to machines to understand why they break

    /

    A startup uses sound to find problems before they start

    Share this story

    There is nothing more frustrating than taking your car into the mechanic with only the vaguest sense that something is wrong. You know that odd little creak or strange whine is new, but you don’t have a clue what it’s trying to tell you. You would like to know before it becomes a serious problem, and an expert is going to charge you a bunch of money to find out.

    What if we could diagnose machines, and keep them healthy, just by listening to the noises they make. That’s the premise of Augury, a startup based in New York City. And since Augury is all about sound, we decided to do this piece as an audio report. Take a listen below and let us know what you think.

    Artificial intelligence has been driving huge advances in the way machines understand sound. The biggest trend at this year’s CES was voice control, specifically the way Amazon’s Alexa had found its way into everything from appliances to robots to automobiles. The success of the Amazon Echo is due in no small part to the fact that computers can now hear and respond to our commands and questions with near human-level accuracy.

    What’s fascinating about deep learning, an approach to AI that has become extremely popular in recent years, is that it can learn to understand patterns and systems without being given explicit instructions or rules by its human creators. That means this new breed of AI can learn to translate between language pairs it’s never been taught, and in the case of Augury, pick up on what our gadgets and appliances are trying to tell us.